Introduction
Current established treatment of hydrocephalus has been surgical, including elimination of its cause or CSF diversion. However, despite technical advancement, failures of CSF diversion are common and approximately 40% children treated with a shunt need to have some intervention or shunt revision within 2 years of the original shunt insertion [
1,
2]. Non-surgical therapies for hydrocephalus have been proposed and practiced so far with limited success [
3]. Therefore, advanced and alternative non-surgical treatments for hydrocephalus are needed.
Aquaporin (AQP) is a water-transporting protein present in plasma membrane and transports water in and out of the cell. AQP1, AQP4, and AQP9 are expressed in brain where they are involved in water homeostasis. AQP4 is the principal, most abundant water channel in the brain providing a low-resistance transcellular clearance route and regulating water flow in and out of the brain. It is concentrated on astrocytic end-feet processes at CSF–brain interfaces (Virchow Robin’s space, glia limitans interna, and glia limitans externa) and at blood–brain interfaces [
4‐
8]. Hydrocephalus upregulates brain water channel AQP4 at the brain–CSF interfaces as well as surrounding astrocyte end-feet and subpial layers in hydrocephalus animals [
9]. The upregulation of AQP4 expression facilitates increased water passage through the brain back into the vasculature [
4]. Thus, AQP4 channels play a role in trans-parenchymal water clearance in obstructive hydrocephalus from the ventricle and extracellular space of the brain into the cerebral vasculature [
10,
11].
AQP4 channels are mainly located on astrocyte end-feet but is later distributed over the whole membrane of astrocytes when hydrocephalus progresses [
12], suggesting that upregulated AQP4 expression is a physiological adaptation to induced hydrocephalus, possibly to aid and facilitate removal of excess water. Water effluxes via glial-vascular pathway or “glymphatic pathway” [
13] by modulating AQP4 channel is an attractive therapeutic target for potential molecular treatment of hydrocephalus.
Erythropoietin (EPO) is an endogenous hematopoietic cytokine, which has anti-inflammatory and angiogenetic effects. EPO binds the EPO receptor, which is widely expressed in the brain [
14]. One of the mechanisms through which EPO reduces brain edema is by regulating expression of the AQP4 water channel. EPO significantly upregulates AQP4 expression in brain tissue and reduces injury in hypoxia [
15] and leads to a better clearance of water excess in brain tissue mediated by AQP4 [
16]. Moreover, EPO can also induce angiogenesis while inhibiting vascular leakage and inflammation [
17].
The role of microRNAs in the expression of AQP4 has recently been studied. Interaction of hsa-miR-668, -1280, -130a, -152, and -939 was noted with the promoter of the AQP4 M1 gene [18. 19]. Among them, miR-130a appears to be a strong suppressor of promoter activity. Previous studies showed that erythropoietin (EPO) delivers protective effects by regulating microRNA (miRNA) expression through their interaction [
20,
21]. Because miR-130a functions as a suppressor that can modulate promoter activity, anti-miR-130a could serve as a potential therapeutic agent in ischemic recovery [
22].
We therefore hypothesize that the severity and duration of hydrocephalus can be rescued by EPO-mediated upregulation of AQP4 by modifying CSF flow into the cerebral blood stream through AQP4 channels. Given the importance of the role of AQP4 in brain edema, we also explored the possibility of identifying miRNAs downstream of EPO signaling that could selectively regulate AQP4 expression.
Material and methods
Animal procedure
Sprague Dawley rats were purchased from Charles River. Female pups of 2 weeks in age with body weight 20–22 g were used in this study. A total of 32 pups were used in this study. Pups were anesthetized by giving isoflurane inhalation. Hydrocephalus was induced in 22 pups through injection of kaolin suspension (50 μl, 10 mg/ml in sterile saline) with 30-gauge syringe into cisterna magna. In control group, pups were injected with 50 μl of saline in cisterna magnum. Kaolin-injected animals were equally divided in to two groups. At the fifth day of kaolin cisternal injection, one group of pups was injected with mouse recombinant EPO (R&D Systems, Minneapolis, MN) 1 μg/pup, intra-peritoneal (i.p.) for 5 consecutive days, while the other was injected with saline for 5 days. All pups were observed throughout recovery from anesthesia and housed in a controlled environment with ad libitum access to food and water. Their behavior and body weight were recorded on daily basis as per Olopade et al. [
23]. At 10th day of post kaolin injection, all pups were sacrificed with institutional guidelines and regulations.
Cell culture
Human astrocytes were obtained from Dr. Rintaro Hashizume, Northwestern University Chicago, IL, USA. Rat choroid plexus epithelial cell line Z310 was provided by Dr. Wei Zheng, Purdue University, West Lafayette, IN, USA. Z310 cells were cultured in RPMI 1640 medium, and the astrocyte cell line was cultured in Eagle’s modified Eagle’s medium. The medium was supplemented with 10% fetal bovine serum, 50 units of penicillin, and 50 μg/ml streptomycin. Human brain microvascular endothelial (HBME) cells were cultured as described previously [
20]. All cells were cultured in T-75 flasks and maintained in a 37 °C incubator with 5% CO
2.
Anti-miRNA transfection
Transfection with anti-miR-668 and scrambled anti-miR miRNA Inhibitor Negative Control #1 from Applied Biosystems was performed using Lipofectamine RNAiMAX transfection reagent (Life Technologies). Cells were seeded to 80% confluency prior to transfection. For the transfection, anti-miRNA-lipid complexes were formed using Opti-MEM medium, Lipofectamine RNAiMAX transfection reagent, and respective miRNA inhibitor. Per well, the final concentration of miRNA inhibitors added amounted to 50 nmol for cells on a six-well plate. After 48 h of transfection, cells were left untreated or treated with EPO for different time points. Total RNAs including miRNA were isolated to examine mRNA expression of AQP-4 and for miRNA quantification. In other experiment, cells were lysed in RIPA buffer for immunoblot analysis of AQP-4 protein expression.
Reverse transcription and quantitative real-time PCR
Reverse transcription followed by real-time quantitative PCR (qRT-PCR) was carried out according to Mohanty et al. [
24]. Quantitation of AQP4 mRNAs was performed using a SYBR Green assay with forward primer, 5-agcctgggatgcaccatca-3, and reverse primer, 5-tgcaatgctgagtccaaagc-3. miRNA from cells was isolated by using the miRNAeasy Mini Kit (Qiagen). Subsequently, cDNA generation was accomplished using the TaqMan MicroRNA Reverse Transcription Kit (Applied Biosystems, Foster City, CA). Following assembly of all components for cDNA synthesis, the reverse transcription reaction was carried out in a thermocycler. TaqMan Universal Master Mix II, no UNG (Applied Biosystems) was used in preparing reactions for qPCR. TaqMan microRNA assay kits from Applied Biosystems were used for miRNA quantified by q-RT-PCR. Data collection for each cycle was done at the completion of the annealing/ extension step. RT primers (for cDNA synthesis) and TM primers (for qPCR) for U6 snRNA control, hsa-miR-130a, hsa-miR-let-7f, hsa-miR-668, and hsa-miR-320a were obtained from Applied Biosystem’s TaqMan. All the real-time PCR experiments were carried out using an Applied Biosystems 7900 sequence detection system.
Western blot analysis
Western blotting using 30 μg of total proteins was carried out as described by Siddiqui et al. [
20]. Membrane was probed with rabbit anti-AQP4 antibody (Abcam) at a concentration of 1 μg/ml in 4% blocking solution. Mouse anti-actin antibody (Santa Cruz Biotechnology) was used as a loading control. The membranes were visualized via enhanced chemiluminescence (Super Signal West, Thermo Scientific).
H&E staining and immunohistochemistry
Pups were perfused with 4% paraformaldehyde and brains were collected and fixed in 10% formaldehyde. After fixation, the brains were embedded in paraffin blocks and sections were prepared (10 μm). Paraffin sections were rehydrated and stained with hematoxylin and eosin dye. For immunohistochemistry, AQP4 protein was probed with mouse anti-AQP4 antibody (dilution 1:200) and with FITC-coupled donkey anti-mouse secondary antibody (Jackson ImmunoResearch, West Grove, PA) (dilution 1:200). DAPI was used as a nuclear stain. Images were viewed at × 10 and × 63 magnifications and analyzed using LSM870 confocal imaging software (Carl Zeiss, Thornwood, NY).
Effect of EPO on AQP4 targeting miRNAs
AQP4 targeting miRNAs were selected with Bioinformatics search in
microRNA.org and Targetscan. We selected miR-130a, miR-320a, miR-let-7f, and miR-668 on the basis of previously published results [
19,
20] to test whether AQP4 is regulated by these miRNAs in response to EPO. Ependymal epithelial cells derived from choroid plexus (Z310) and astrocytes were treated with EPO (20 IU/ml) for different time points. Subsequently, RNA was isolated and miR-130a, miR-320a, miR-let-7f, and miR-668 levels were determined by qPCR.
Cultured cerebrovascular endothelial cells (ECs) were studied for the effect of EPO on AQP4 expression and angiogenetic effects of EPO on ECs. HBME cells (5 × 104) were suspended in 500 μl of serum-free DMEM media with or without EPO 20 IU/ml on a 24-well plate pre-coated with reduced factor Matrigel (Corning, NY). After 6 h of incubation at 37 °C and 5% CO2, pictures of the vascular tube-like structures were taken under × 20 magnifications. The number of polygonal areas formed by these tube-like structures was counted for each field.
Statistical analysis
Comparison between different pair was done using Student’s t test. Comparison of three or more groups was performed using one-way analysis of variance (ANOVA). A P value of 0.05 or less was considered to be statistically significant.
Discussion
Present study examines the role for EPO in modulating AQP4 water channels in experimental hydrocephalus. It is noted that EPO treatment upregulates AQP4 expression and reduces dilated cerebral ventricles in kaolin-induced model of obstructive hydrocephalus in rat pups. Also, we identified the role of miR-668 as an endogenous modulator of AQP4 expression in response to EPO.
Earlier studies by Tourdias et al. showed an upregulation of AQP4 in astrocytes in response to communicating hydrocephalus, which may participate in clearance of excess fluid to the blood stream [
12]. On the other hand, Mao et al., who studied effects of obstructive hydrocephalus on AQP4 expression in rat, found changes in mRNA level but not in protein level of AQP4 channel [
4]. In our study, we did not observe substantial difference in AQP4 expression between normal and obstructive hydrocephalic brain. This observation suggests that differences in AQP4 expression might depend on the model and severity of hydrocephalus.
Malformation of ependymal epithelial cells has been implicated in the etiology of hydrocephalus [
1,
25]. Consistent with these observations, we found denudation of ependymal structure after kaolin injection, which dramatically reduced after EPO treatment likely due to decreased intracranial pressure (ICP) and ventriculomegaly. Thus, the ependymal layer denudation is the result of pressure hydrocephalus and appears to be reversible, parallel to the reduction in ventricular size following the EPO treatment. We found an increased expression of AQP4 in the ependymal layer in vivo as well as cultured astrocytes following EPO treatment. Therefore, upregulation of AQP4 expression in the ventricle ependymal layer could contribute to clearing of CSF out of the ventricle into parenchyma. Since AQP4 proteins are also located at astrocyte end-feet, we hypothesized that their upregulation could facilitate the removal of excessive water from the parenchyma into the blood thereby promoting EPO-mediated CSF absorption. This in turn would reduce hydrocephalus by upregulating AQP4 expression at major junctions of fluid compartments. Our findings are in line with previous observation showing that EPO-mediated increase in expression around hematoma helps in reducing brain swelling [
16].
In this study, we demonstrate that EPO decreases miR-130a and increases miR-668 which in turn upregulates AQP4 expression in cultured ependymal epithelial cells and astrocytes. Inhibition of miR-668 by anti-miR-668 prevents the EPO-mediated upregulation of AQP4. AQP4 has been shown to be a direct target of miR-320a and miR-130a and under ischemic condition, delivery of anti-miR-320a and anti-miR-130a upregulates AQP4 expression and exhibits beneficiary effect on infarct size [
18,
19]. However, we observed miR-668 as a strong activator of AQP4 expression in response to EPO. This positive regulation of AQP4 expression by miR-668 may be oversimplistic, since other signaling molecules down to erythropoietin receptor (EPOR) may be involved in AQP4 expression [
26,
27]. Further investigation is needed of the roles of microRNA in hydrocephalus physiology and management.
Endothelium transports water from peri-vascular astrocyte end-feet into vessel lumen by diffusion and vesicles mediated transport. Pro-angiogenetic property of EPO has been well documented [
28]. In our experiment, when cultured HBME cells treated with EPO in in vitro, we did not observe any changes in AQP4 expression in endothelial cells but noted increased endothelial tube formation. Therefore, we hypothesized that increased number of vessels may potentiate the exit of water from peri-vascular astrocyte end-feet into blood vessel.
Compliance with ethical standards
Open Access This article is distributed under the terms of the Creative Commons Attribution 4.0 International License (http://creativecommons.org/licenses/by/4.0/), which permits unrestricted use, distribution, and reproduction in any medium, provided you give appropriate credit to the original author(s) and the source, provide a link to the Creative Commons license, and indicate if changes were made.